It is all too easy to criticize contemporary social psychology. Its obsession with methodological exactitude and its craving for scientific status have produced a discipline which at first sight appears to be devoted to the detailed analysis of insignificant phenomena. Disgruntled polemics, however, do not uncover the roots of the problem, nor do they propose solutions. Suggestions for new directions should not be accepted uncritically on the grounds that any change must by Three of the major alternatives to current mainstream social psychology are ethnomethodology, symbolic interactionism and ethogeny. These approaches will be called collectively 'the new social psychology'. O f the three, only ethogeny is specifically addressed to problems in social psychology ( H a d and Secord, 1972; H a d , 1972 and. Symbolic interactionism aims to occupy a middle ground between social psychology and sociology (Blumer, 1969), in order to develop a 'sociological social psychology ' (Lindesmith, Strauss and Denzin, 1975 : 7). Ethnomethodologists seem to direct their attention to problems within sociology (e.g., Garfinkel, 1967; Zimmerman and Pollner, 1971); however, it will be suggested that their approach has much in common with symbolic interactionism and ethogeny. Moreover, the ethnomethodological approach has been used directly as a critique of orthodox social psychological methodology (Heritage, 1974). More generally, since these new approaches attempt to challenge existing academic disciplines, it would be unfair to categorize them in terms of the traditions they seek to replace.The advocates of the new social psychology see themselves as revolutionizing the study of social behaviour. Ham6 and Secord call for a 'paradigm-shift' in psychology, hoping themselves 'to articulate the New Paradigm ' (1972: 19). In a similar spirit Zimmerman and Pollner (1971: 93) talk of a 'paradigmatic shift ', and Zimmerman and Wieder (1971: 294) refer to the 'radical character of the ethnomethodological enterprise'. Broadly speaking, all three approaches oppose a positivistic science of social behaviour. Blumer (1969) criticizes what he calls 'variable analysis' and those social psychologists who are 'fortified by an oversimple view of scientific procedure which would reduce the scientific act to a search for quantitative information and quantitative relations' ([p.] 72). Ham6 (1974) advocates that social psychologists should cease attempting to predict and control behaviour but should aim for understanding. Similarly, Silverman (1972a: 169) states that ethnomethodologists recommend a different form of investigation than 'happily' relating A to B.A mechanistic model of man is rejected. Garfinkel (1967: 68) refers to the model of man produced by orthodox sociology as being a 'cultural dope who produces the stable features of the society by acting in compliance with preestablished and legitimate alternatives of action that the common culture provides'. In the same vein, there is the 'psychological dope' as portrayed by models which ...